Preprints of the Metadiversity Conference Proceedings

  Session 6: The Metadata Challenge for Museums

Museum Metadata: Who Pays?

RAY LESTER, Head of Library & Information Services, Natural History Museum, London

In this Paper I have tried to relate in as simple a fashion as possible key themes discussed in the last few days to the reality of work in London, UK. The Natural History Museum is a reasonably significant institution in global terms:

• 68,000,000 biological/mineralogical specimens
   
• 1,000,000 library volumes
   
• 10,000 current serials
   
• 500,000 art and other information artifacts
   
• 300 scientists
   
• 30 librarians

And, as has been commented here for other large species- and specimen-based collections, the data relating to identification, location, habitat and so on, which can be digitally captured from the collections, converted into information, and delivered via the Web, is potentially of value to people worldwide, both in the developed and developing countries. That information, and its underlying "rich data"–the word "data" used of course here to encompass text, image, moving image, voice, and multimedia combinations of such–can be discovered and made accessible to the customer via meta-information and meta-data.

The Museum wishes to provide such electronically mediated access to representations of its library, art and archive holdings, as well as to those of its specimen collections, especially in the senses that "our" metadata–by "our" I mean metadata created by people employed by The Natural History Museum–is likely to be contributed in the future to what I term here Web Consortia. (The Museum for instance is already committed to participate in Species 2000 and is a sponsoring partner of the UK's proposed National Biodiversity Network.) Also, in creating "our" metadata, we might wish to make use of–query, make "fair use" of, in an intellectual property sense?– metadata stored in the print volumes of the Museum's Library (such as the scientific journals, and the abstracting and indexing volumes). In those contexts, I thought that it might potentially be useful to ask in this forum, at this juncture:

Who pays for metadata?

It will not surprise you that I do not expect here an answer to the question, though it would be nice if my comments stimulated some discussion. The question itself is obviously rather fundamental, and it would need more time and energy than we all have at this point in the conference to do it justice. Cliff Lynch in some comments earlier today eloquently elaborated some of the facets that would need to be addressed in producing anything approaching a comprehensive answer to the question. So my prime intention in this Paper–using The Natural History Museum as an exemplar–is to try to explain why we believe in London that the question needs to be answered–and moreover, to be answered through an international cooperative effort of the types of stakeholders represented at this conference. Once, with our now many and various "metadiversity" networks, we move from research grants and lottery funds and politically inspired initiatives to doing things "for real," we will need to put in place global mechanisms for deciding who pays for the metadata (and metainformation) used in the networks. The participating stakeholders will positively need to do that, because we are operating in a non-market economy or, at least, in a mixed economy. I presume none of the stakeholders who work outside of the market sector would wish the distribution of metadiversity information wholly to be market driven.

The Natural History Museum has a vision of its data and information flowing "freely" around the Internet. That vision is fully compatible with the Museum's overall mission:

"The Museum's mission is to maintain and develop its collections and use them to promote the discovery, understanding, responsible use and enjoyment of the natural world"

and with the mission statement of my own Museum Department:

"Organising, preserving and communicating knowledge of the natural world."

The vision is to make all "our" natural world data, information, metadata and metainformation available directly or indirectly to customers via the World Wide Web. We wish our offerings to be seamlessly entwined with the offerings of other compatible institutions worldwide. The Director of the Museum–as all such Directors!–would like this to happen next week, including extending our offerings to include tacit taxonomic knowledge and perhaps even the odd bit of wisdom!

There is a phrase often used in computing circles–"managing expectations." I guess one has not done that very well. We have painted a vision of using the wondrous technology to deliver data and information at all levels of granularity, in all media, to customers located in other institutions, in other professional domains, in other funding sectors, and in other political jurisdictions. But we have not said sufficiently is, "This is going to cost a great deal of money, and there are a number of critical global policy decisions that will need to be made before the vision can be realised, including a determination of who pays."

Naturally, when one starts to disentangle the policy decisions that need to be made, one finds that the necessary frameworks are all there in the literature–albeit framed at other times and in other milieu. A good example is the taxonomy represented in this Table, taken from Taylor's 1986 seminal text Value Adding Processes in Information Systems (Taylor 1986):
 

To ensure a productive debate leading to a single, agreed-upon, global, taxonomically-based information architecture acceptable to all significant stakeholders (and there has to be one seamless over-arching architecture–which is, of course, not the same as saying that there needs to be one over-arching global information system), we need to find the time and energy to work through the details, using agreed frameworks such as Taylor's.

I am fond of using the phrase "The Middle" to denote all those building blocks within the overall information architecture that are outside the direct management and command of either the resource-holding institution–in this case the Natural History Museum–or of the organisation, if any, wherein the customer for such held resources resides. In the widest sense, value-adding processes have to occur if the "rich data" captured and sitting on computer servers within my own and other similar resource-holding institutions is to be transformed into information which can be used by the remote customer for the task in hand. That value-adding clearly can occur:

• within the Natural History Museum (or other resource-holding institution)
   
• within the customer’s own organization
   
• in "The Middle," overseen by an entity, or entities, quite separate from the Museum or the user and his/her parent organization

I am calling these "middle" entities here Web Consortia.

The overall network architectural infrastructure–or "superstructure"–has to characterise and maintain each of a number of types of building blocks: the building blocks of a particular type will occur within the resource-holding institution, within the customer organisation, or in "The Middle." These building blocks can be characterised as:

• information technology hardware/software
   
• information systems standards/protocols
   
• metadata
   
• rules and regulations of system use
   
• value-adding intermediaries, where needed
   
• coordinating bodies

A building block that is not optimally available and maintained will hinder, and may even prevent, the communication of a server-based resource to the customer for that resource. When we come to do this "for real," we cannot have the server going down, the wrong Z39.50 profile being used, the species name being entered incorrectly, the user un-authenticated, the customer interface being clumsy and leading to the user giving up, or the coordinating-body Help Desk being closed. And all such elements and much, much, more need management.

In an important book published earlier this year (Weill & Broadbent 1998), Marianne Broadbent and Peter Weill reported on some detailed empirical research carried out over a number of years. The question addressed was, "How is it that some large, multinational enterprises succeed in leveraging IT, whilst others fail?" In other words, what are the secrets of getting freestanding companies, albeit part of the same corporate conglomerate (and remember as regards "The Middle" in my conception, we have not even got that potential sanction) to work together and use global IT networks to deliver real cost/benefit?

I asked Marianne Broadbent, shortly after this book was published, what she would advise for Web Consortia of the type discussed in this Conference. Her recommendations were:

• Must have commitment from stakeholders at the level of CEO/The Board
   
• Must relate to maxims in each stakeholder that are strong and about the core roles of each parent organisation
   
• Must focus on the "have to" rather than the "nice to have"
   
• Must be able to display demonstrable benefits
   
• Must be a long-term, sustainable process
   
• Someone must have a remit to command the whole information system

The key recommendation, it seems to me, is the last. Someone has to be in overall command where, as noted earlier, the market-place is not wished to be the ultimate arbiter of who supplies and gets what globally dispersed metadiversity data and information.

And I suggest that such will need to be the case even when we do, indeed, agree on a single species name to link together data and information held in dispersed and disparate information systems. In fact, one critical element of the commitment needed from the participating stakeholders in Web Consortia is indeed to use that agreed-upon single-species name internally as a pointer to the other species and related specimen data and information each stakeholder holds and wishes potentially to make available to external customers. Note in passing that it has to be that way around. The model has to be the generation by the rich data-holding institution of a Dublin Core–or a Darwin Core or whatever–record containing the agreed-upon species name.

This speaker is in theory–in theory!–commanding the resources needed to deliver to the outside world this necessary metadata and metainformation about the resources held in the vaults of the Natural History Museum. But this speaker is in no sense in command of the Web Consortia via which such data and information might eventually be delivered to external customers. Even within the Museum, one's theoretically commanding role has to take account of the views of our friends in the Museum Department of Exhibitions and Education, who are much exercised by matters of corporate identity, and by our other friends in the Department of Development and Marketing, who similarly are exercised by the need to raise income for the Museum over and above that which arrives courtesy of the UK Government's grant-in-aid. And then, in the centre of all this, we have the information and communication needs of the Museum scientists themselves to address.

There is such a momentum now within the Natural History Museum to give Web access to the Museum's cornucopia of resources ("increasing access" being also a favourite theme of the Blair government) that I have been able to propose to the Museum the creation by my Department of a full-fledged Business Plan with costed options designed to achieve this. (The Museum's Director likes business plans with costed options!) In preparing this Business Plan, the environmental assumptions we would make are:

1. There is indeed customer demand for Web-based information relating to Museum "rich data" and made accessible via perusal of "metadata" (and "metainformation").
    
2. In making such Museum information electronically available, the most important factors to get right are quality, presentation, and indexing.
    
3. The prime strategic decision for the Museum is the extent to which it will participate in future Web Consortia, and whether as secondary, or as lead, partner.

"Web Consortia" are conceived as trusted gateways to the "best" pockets of data/information/metadata/metainformation accessible in total via the World Wide Web. Such quality gateways will gradually replace the generic search services such as the present Yahoo and HotBot for purposeful enquirers who do not know where in the world the reliable information (and data) they need resides. Given the universal nature both of the underlying Web technology and of the Museum's topical focus ("the natural world"), it is felt that it is inevitable that there will emerge for the Museum's subject field a relatively small number of Web Consortia "brand leaders." These organisations, which could be located anywhere in the world, will become the preferred first ports-of-call when customers wish to access and use data/information/metadata/metainformation about the "natural world" via the Web but do not know where trustworthy elements of such digital representations can be found.

The proposed study that would lead to the proposed Business Plan with costed options can be summarised:

Test environmental assumptions
Audit of present Museum capabilities
            IT infrastructure
            Staff for resource creation and maintenance, market interaction
            and
            management/administration
Customer market analysis
            Science
            Policy
            Public
Competitor analysis
            Providers of data/information/metadata/metainformation
Web consortia
Prospective Museum competitive advantage
Marginal resources required to fulfill target market needs
Revenue implications

In conclusion, it is now easy to see why the Natural History Museum must determine an answer to the question, "Who pays for the metadata?" Answering the question for any specific object stored within the Museum is not, however, easy. Copies of most objects can be delivered in a variety of digital–and other–forms, each of which will need its own metadata. Such metadata–especially if the Museum is going to consider charging for access to the underlying object surrogate–will need to go beyond descriptive metadata, and include "rights" (and instantiation and administrative) metadata. The Museum will also need to recognise the different arenas wherein the metadata will be needed and will be used, e.g., at the point of resource creation, when it is made available, when it is used. And the Museum will need finally to recognise that even when it does decide that needed metadata must be paid for by the Museum, or by the customer, deciding how the former should be costed (or priced) and the latter should be priced (or costed) brings in train a whole new set of challenges–given the underlying economics of information, with its non-depletability, its potential non-excludability, and its almost-always positive externalities. To show to you a slide by McGillivray of a wonderful image of a heron stored in our art collections did not "cost" me anything. Or did it?
 

References

Taylor, R. S. 1986. Value-adding processes in information systems. Ablex, Norwood, NJ.
Weill, P. and M. Broadbent. 1998. Leveraging the NEW infrastructure. Harvard Business School Press, Boston, MA.

Previous | Next

Copyright © 2003 NFAIS. All rights reserved. No part of this product or service may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written consent.

Privacy Policy